Campylobacter fetus is a pathogen of humans and animals. All wildtype C. fetus strains possess regular, paracrystalline high molecular weight (97, 127, or 149kDa) surface layer proteins (SLPs) that are critical for virulence. The SLPs inhibit complement binding, which leads to both serum and phagocytosis resistance, and they undergo antigenic variation. C. fetus cells possess 8 or 9 homologs of sapA, which encodes a 97 kDa SLP. The homologs share multiple areas of homology with each other, are clustered on the bacterial genome, and together with an invertible region (IR) of 6kb form the sap locus. The investigators previously showed that the DNA rearrangement (inversion) plays a major (if not exclusive) role in the antigenic variation, and that mutation of recA removed all detectable variation. They now have developed animal models to examine SLP variation, have shown that antigenic variation due to sap locus recombination occurs in recA strains, and have moved the IR onto the E. coli chromosome. The first aim of the proposed work is to identify the recombination pathways that permit the sap locus inversion. Their hypothesis is that the inversion depends on conserved recombination genes including those in the RecBCD, RecE, and RecF pathways. Alternatively, they might find that C. fetus has species-specific recombination pathways. The ultimate goal will be to define and mutate the recombination genes in C. fetus and examine the phenotypes of mutants. Their second aim is to define the structural features present in the sapA homologs that are required for recombination to occur. They propose to complete the map of the sap locus and to create mutants that will permit assessment of the necessary structural features. As part of this goal, they plan to identify the sites of double strand breaks that are the initial sites for DNA strand exchange. The third aim is to examine sap rearrangement in vivo in mice. After establishing the model, they can examine the role of host responses in directing the evolution of antigenic variation, and the fitness of defined mutants. Finally, the data generated can be used to develop a mathematical model of antigenic variation that focuses on the relationship between microbial recombination generating diversity and host selection determining the direction of change.